Beyond algorithms: how folk theories of information homogeneity differ from academic theories across the contexts of the US and China

While previous studies have examined online homogenous information experiences as theorized by various concepts (e.g., echo chambers and filter bubbles), we know less about how the general public makes sense of these encounters; their folk theories could be very different behind the same behaviors. Meanwhile, the contextual factors that shape the configurations of folk theories remain underexamined. This study thus leveraged a comparative approach, which contrasted folk theories with formal academic theories and compared folk theories across contexts. Combining large-scale computational frame analyses of social media posts (Weibo: N = 245,244; Twitter (now X): N = 171,125) and qualitative textual analysis, our mixed-method approach identified seven folk theories. They were compared with theories extracted based on a content analysis of relevant academic papers (N = 132). The cross-context comparison shows that unlike academic studies that emphasize algorithms and platforms’ roles, laypeople primarily attribute information homogeneity to political actors (US) and interest-based groups (China). This study highlights folk theories’ contextual contingencies and suggests a comprehensive theoretical framework that explains the configurations of folk theories in the current high-choice media environment.